Our Center focuses on factors affecting outcome in cognitive behavioral therapy (CBT) in anxious youth. CBT is effective for 50-60% of anxious youth. Because mechanisms of recovery are poorly understood it is difficult to predict whom CBT will be effective for or how to refine it. Initial research has identified abnormalities in biological substrates of emotional information processing likely targeted by CBT, specifically, a pattern of excessive vigilance for threatening information followed by blunted sustained emotional reactions. This pattern is likely subserved by the amygdala and prefrontal regions. Understanding whether these mechanisms are targeted in CBT will 1) allow prediction of whether emotional information processing styles predispose individuals to recovery in CBT and 2) refinement of CBT to more specifically target brain mechanisms. This project requests funds to assess 200 anxious and 40 control youths' emotional information processing using functional magnetic resonance imaging (fMRI), event-related potentials (ERP), and pupilometry/eyetracking. Anxious youths will receive 16 sessions of CBT or an active control intervention (child centered therapy). All participants will be assessed every 2-4 weeks over the 14 week intervention period with pupilometry/eyetracking. After session 16, participants will again be assessed on the same tasks using all our measures. Recovery in CBT is expected to be strongest for anxious youth with heightened early limbic responses, which are targeted by CBT. CBT is expected to decrease initial limbic responses to emotional stimuli, and increase sustained responses (e.g., decrease avoidance). Positive results could link brain function to recovery, allowing prediction of which anxious participants will recover in CBT, illustrate mechanisms by which CBT leads to recovery, and could validate proposed mechanisms of action for CBT. Negative results would suggest that recovery in CBT may depend on different mechanisms of action than those which have been proposed by the clinical and affective neuroscience community. The scale of the proposed research is unparalleled in examinations of neural correlates of emotional information processing in anxious individuals, particularly with developmental populations and examinations of recovery. This magnitude is necessary to understand factors underlying the considerable heterogeneity in treatment response in this population. Our group is uniquely suited to carrying out the proposed large-scale research.